Deep-Sea Research I 48 (2001) 1325}1346
Variations in deep-sea hydrothermal vent communities
on the Mid-Atlantic Ridge near the Azores plateau
D. Desbruyères *, M. Biscoito, J.-C. Caprais , A. Colac7 o, T. Comtet,
P. Crassous , Y. Fouquet , A. Khripouno! , N. Le Bris , K. Olu , R. Riso,
P.-M. Sarradin , M. Segonzac , A. Vangriesheim
Centre de Brest de l+IFREMER, BP 70, 29280 Plouzane& , France
IMAR } Museu Municipal do Funchal (Histo& ria Natural), Rua da Mouraria, 31, 9004-546 Funchal, Madeira, Portugal
IMAR } Laborato& rio Marn& timo da Guia, Faculdade de CieL ncias de Lisboa, Estrada do Guincho, 2750 Cascais, Portugal
Plymouth Marine Laboratory, Citadell Hill, Plymouth PL1 2PB, UK
UFR Sciences et Techniques, 6 av. Le Gorgeu, B.P. 809, 29285 Brest, France
Received 9 August 1999; received in revised form 24 May 2000; accepted 31 July 2000
Abstract
Near the Azores Triple Junction as the Azores Plateau is approached, the ridge axis becomes shallower; its depth
decreases from ca. 2400 m in the RAINBOW vent "eld (36313N) to ca. 850 m in the MENEZ GWEN vent "eld (37335N). In
this area, extensive mussel beds of the mytilid Bathymodiolus azoricus dominate the hydrothermal vent fauna, along with
populations of three shrimps (Rimicaris exoculata, Mirocaris fortunata and Chorocaris chacei). The main physical and
chemical characteristics of the vent habitat were studied by discrete sampling, in situ analysis and sediment trap
moorings. The vent fauna is distributed along a variable band where the vent #uids and seawater mix, with R. exoculata
living in the most concentrated areas and Bathymodiolus azoricus in the most diluted zones. Various non-endemic
species live at the border of the vent "eld. The variations observed in structure and composition of the communities
along the depth gradient are most likely due to changes in vent #uid toxicity (metallic and sulphide content) and
suspended mineral particles, which render the #uids harsher for species living there. The main faunal di!erences observed
between LUCKY STRIKE and MENEZ GWEN hydrothermal "elds are due to an impoverishment in the hydrothermal
endemic species and to the penetration of bathyal species. The comparison of the three studied vent "elds suggests the
existence of a succession of several biogeographic islands rather than a single province. 2001 Published by Elsevier
Science Ltd.
Keywords: Deep-sea; Hydrothermal vents; Mid-Atlantic Ridge; Azores Triple Junction; Ecosystems; Habitat characterisation
* Corresponding author. Fax: #33-2-98-22-4757.
E-mail address: ddesbruy@ifremer.fr (D. Desbruyères).
0967-0637/01/$ - see front matter 2001 Published by Elsevier Science Ltd.
PII: S 0 9 6 7 - 0 6 3 7 ( 0 0 ) 0 0 0 8 3 - 2
1326
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1. Introduction
The discovery of hydrothermal vents on ocean ridge crests occurred in 1977, o! the Galapagos
Islands (Lonsdale, 1977), but the exploration of the mid-Atlantic Ridge only started in 1985 (see
Van Dover (1995) for details). Several ecological studies (Segonzac, 1992; Van Dover, o.c.) on the
two main deep hydrothermal vent "elds, TAG (263N) and Snake Pit (233N) ca. 3500 m deep,
showed faunal assemblages very di!erent from those found on the East Paci"c Rise. Caridean
shrimps, in particular Rimicaris exoculata, dominate populations in these deep Atlantic "elds.
Conversely, alvinellids and tube-worms are not observed.
In 1992, during the US mission FAZAR, a new site was discovered at 37318N, 1700 m deep
(LUCKY STRIKE). A preliminary study of the communities' composition at this site was done during
six dives of the US deep research submersible ALVIN and suggested the existence of two distinct
biogeographical provinces on the Mid-Atlantic Ridge (Van Dover et al., 1993). Following the "rst
discovery, the European Union programme MAST 2 * MARFLUX/ATJ, the diving cruises of
the French submersible NAUTILE, the MAST 3 AMORES programme and the PortugueseFrench SALDANHA cruise, continued and completed the exploration of the Azores Triple
Junction area, establishing its depth limits. From 1993 to 1998, the hydrothermal vent communities
thriving in three major vent "elds located on three segments of the southeastern limb of the Azores
Triple Junction (RAINBOW, LUCKY STRIKE and MENEZ GWEN) within 23 of latitude, were explored and
studied.
The deep-sea hydrothermal vent community composition and structure are a!ected (1) by
linking and isolating mechanisms between vent "elds (Hessler and Lonsdale, 1991; Tunnicli!e,
1991), (2) by local conditions (chemistry and particle content of #uids and substratum patterns)
(Johnson et al., 1988), and (3) by instability of venting, which induces an extinction-colonization
dynamics (Cann et al., 1994; Desbruyères, 1998; ChevaldonneH et al., 1997). In the Azores Triple
Junction area, as the Azores Plateau is approached, the ridge axis gets shallower; its depth
decreases from ca. 2400 m in the RAINBOW vent "eld (36313N) to ca. 850 m in the MENEZ GWEN vent
"eld (37335N). Because of phase separation processes, the accompanying drop in hydrostatic
pressure directly a!ects the chemistry of the high-temperature e%uents and their mineral particle
content. Source rocks hosting the hydrothermal convection also a!ect the chemistry of the e%uents
and deposits. The RAINBOW ultrama"c hosted system produces acidic #uids enriched in metals and
gases (H and CH ) and relatively impoverished in H S (Charlou et al., 1997). The physical settings
of the area studied provide a unique opportunity to describe the in#uence of a wide range of
chemical and physical factors on the composition, distribution and dynamics of vent communities
at slow spreading ridges when geographical isolation by transform faults is absent or reduced.
Lucky Strike Cruise, May}June 1993, Chief-Scientist, Charles Langmuir.
Coordinator Henry Bougault, IFREMER, France.
DIVA 1 Cruise, May 1994, Chief-Scientist Yves Fouquet and DIVA 2 Cruise, June}July 1994, Chief-Scientists Daniel
Desbruyères and Anne-Marie Alayse.
FLORES Cruise, July}August 1997, Chief-Scientist Yves Fouquet; MARVEL Cruise, August}September 1997,
Chief-Scientists Daniel Desbruyères and Anne-Marie Alayse and PICO Cruise, June}July 1998, Chief Scientist and
Programme Coordinator, Daniel Desbruyères.
July 1998, Chief-Scientist Fernando Barriga.
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1327
2. Materials and methods
2.1. Animal collection, mapping and microdistribution
The present study is based on in situ observation with the submersible NAUTILE, video record
analyses and fauna sampling during dives. Data are provided by four cruises, in summer 1994
(DIVA2), 1997 (MARVEL) and 1998 (PICO & Victor Première). Mapping of vent sites was done
with NAUTILE's acoustic navigation records on high-resolution EM12 bathymetric maps (from
FLORES cruise), and the GIS analysis ArcView威 software. At selected hydrothermal sites, video
analysis was used to describe the microdistribution of each species and its relationship with
substratum and #uid escaping structures.
2.2. Chemical and thermal characterisation of the habitats
Fluid samples were obtained with two samplers manipulated by the submersible NAUTILE: 750 ml
titanium syringes (Von Damm et al., 1983) associated with an autonomous temperature probe for
the hot #uids and a new multisampler unit using four 200 ml titanium bottles associated with an
autonomous temperature probe. Sampling was done between and inside the populations marked
with quadrats 0.5;0.5 m in size as topographic benchmarks. Water sampling locations were
chosen in relation to faunal assemblages and microenvironments: (1) #uid di!usion or vent
proximity, (2) within R. exoculata swarms, (3) inside Mirocaris fortunata populations, (4) on
mussel-beds (small- or large-sized) and (5) associated or not with shrimps (M. fortunata and
Chorocaris chacei).
During the `VICTOR PREMIEE REa cruise, a new chemical analyser (Le Bris et al., 2000) was
successfully tested in the LUCKY STRIKE area, and a series of in situ S (total dissolved sulphide),
nitrate and temperature data were obtained. This in situ analyser, called ALCHIMIST, is based on
#ow injection analysis and colorimetric detection. Sample processing and analytical methods are
displayed in Table 1.
Discrete temperature measurements were made with the probe operated by the submersible, and
temperatures over a time series were obtained at di!erent population beds with HOBO威
and MICREL威 probes (Fornari et al., 1994). Calibration was done with a standard reversing
thermometer.
2.3. Particulate yuxes
At MENEZ GWEN and RAINBOW (`Flores 5a vent) vent "elds, triple sediment traps (see Khripouno!
and AlbeH ric (1991) for description) were deployed by the submersible NAUTILE during the
MARVEL cruise for 10 and 16 days, respectively. On both occasions the trap was set close to an
active vent. At LUCKY STRIKE vent "eld, a triple trap was deployed 1.5 m south of a smoker called
SINTRA during the DIVA 2 cruise. In order to obtain a pelagic #ux reference, another triple trap was
also moored during the MARVEL cruise outside MENEZ GWEN hydrothermal "eld. Prior to
deployment, the sampling bottles were "lled with "ltered seawater. No poison was added because
bacterial decomposition was considered to be negligible during the mooring time (2;5 days at
MENEZ GWEN, 4;4 days at RAINBOW and 1;25 days at LUCKY STRIKE. After recovery, samples were
1328
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Table 1
Sample treatment and analytical methods used in the study of microenvironments
Parameter
Processing
Analytical method
Remarks
pH
Analysis on board
253C TRIS bu!er
S (H S#HS\)
Analysis on board
Potentiometry electrode for
sulphide-rich medium
Colorimetry (Fonselius, 1983)
CO , CH
Analysis on board
Nutrients
Freezing
Cu, Pb total dissolvable metal HNO suprapur威, 20 l in
20 ml, ambient T
S and NO\#NO\
In situ analysis ALCHIMIST
Head space-GC-HWD FID
(Sarradin and Caprais, 1996)
Colorimetry, segmented #ow
Potentiometric stripping
analysis (Riso et al., 1997)
FIA (Le Bris et al., 2000)
DL "0.5 mol l\
rsd"5%
10%
rsd"5%
Cu rsd"5%
Pb rsd"5%
1.5%
DL"detection limit.
NH>, NO\#NO\.
stored in the dark at 43C pending analyses. In the laboratory, macro- and meio-organisms were
removed from the particle samples. Samples were then rinsed with Milli-Q puri"ed freshwater
(pH&7), freeze-dried and weighed. Total carbon and nitrogen contents were measured in duplicate with a Carlo-Erba威 NA 1500 auto-analyser. Organic carbon content was measured with
a Leco威 WR12 elemental analyser after removal of carbonates with a 2 N HCl solution (Weliky et
al., 1983). Inorganic carbon content was calculated as the di!erence between total and organic
carbon contents. Total sulphur was determined with a Leco威 CS-125 auto-analyser. Elemental
analysis of particles was undertaken by EDAX威 l DX-4i X-ray spectrometry. Standards were
prepared in the laboratory from pure chemical compounds. The average accuracy of the analyses
was 15%.
3. Results
3.1. Physical settings and faunal characteristics
From north to south, three distinct vent "elds were visited (Fig. 1).
MENEZ GWEN is situated in the volcanic segment between 37335N and 383N. The main system is
formed by a 700 m high volcano, 17 km in diameter (Fouquet et al., 2000). A 2 km wide axial graben
splits its summit. Several active sites were located on the southeast and east slopes of one small
volcano growing at the northern end of the bottom of the graben (Fig. 2) at depths ranging from
840 to 865 m. At `PP 10/F 11a sites hydrothermal precipitates cover an area of about 50 m in
diameter. Anhydrite chimneys up to 2 m high are present at the summit of this low elevated
hydrothermal mound. A few patches (500 cm each) of mussels were present. A few geryonid crabs
(Chaceon aznis) were present in the vicinity. A second and more important site (markers D 9, PP
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1329
Fig. 1. General location map of the eight active hydrothermal vent "elds known to date on the Mid-Atlantic Ridge
between Equator and the Azores Archipelago. The major transform faults, which constitute the principal isolation
structures, are also drawn on the map.
11, F 12) is located in an escarpment on the slope, between 860 and 842 m depth. It is bordered in
the upper part by a "eld of pillow lava and laterally by crumbled rocks. Hydrothermal deposits,
from which a 10}403C refringent #uid di!uses, occupy the centre of the site. An active chimney
situated on the northwest escarpment belched out #uid at 2773C. The site's periphery was occupied
by a little dense belt of hydroids. Numerous bathyal "shes have been found around the site:
Chaunax sp., Trachyscorpia cristulata echinata, Neocyttus helgae, Epigonus telescopus and Beryx
splendens (Saldanha and Biscoito, 1997). In the limit between lava and the anhydrite deposits, there
were important mussel' colonies. Specimens sampled showed a maximum shell length of 111 mm.
There were no commensal scale-worms inside these mussels. Numerous patelliform gastropods
were present on the mussels' shells, in particular Protolira valvatoides and one new species of the
genus Lepetodrilus. These mytilid populations consisted mainly of large individuals (l'40 mm),
but the two younger cohorts (modes 4 and 18 mm) were also present and represented ca. 20% of the
population (Comtet and Desbruyères, 1998). Extensive bacterial mats covered some of these
populations. On the active deposits (chimney walls) and amongst mussels, important populations
of C. chacei and M. fortunata were found. S. mesatlantica was the dominant indigenous predator,
while many C. aznis came to feed on the mussels.
1330
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Fig. 2. Map of the MENEZ GWEN vent "eld (`EM 12a bathymetric survey * FLORES cruise). The small (young) volcano
is growing on the bottom of the axial graben, which split an 800-m tall circular volcano situated at the centre of the
volcanic segment adjacent to LUCKY STRIKE segment.
At the PP 32/33 markers, the faunal characteristics were similar to the above. Mussel populations were very important and covered virtually all-available rock surfaces. In contrast with
previously studied sites, the size of mussels in this population was heterogeneous. Some mussels
carried the commensal Branchipolynoe seepensis. Numerous gastropods were present on the
mussels. Geryonid population around and inside the site was important. A single "sh of the genus
Gaidropsarus was observed inside a crevice amongst mussels.
LUCKY STRIKE is one of the largest known active "elds in the modern ocean. The hydrothermal
#uid, with a temperature ranging between 170 and 3243C, has #uid characteristics (temperature,
chlorinity and gas concentration) varying from site to site within the "eld (Charlou et al., 2000a,b).
The hydrothermal vent sites of this "eld are distributed around a lava lake (Fig. 3), in particular in
the southeastern and northwestern zones. Well-de"ned active chimneys such as Ei!el Tower, Y3 or
Elizabeth, belching out very hot #uids and zones where hydrothermal activity is more di!use can
both be found at LUCKY STRIKE. For practical reasons, an equivalence of vent site names given
during French and American cruises corresponding to passive markers is proposed (Table 2).
The fauna is described from the biggest and most studied site EIFFEL TOWER considered as
example of the Lucky Strike "eld. In fact, no noticeable di!erence was observed in the dominant
species composition or microdistribution except at the Y3 site. Except for the #aky anhydriteand barite-clad steep slopes, which were inhabited by more or less extensive populations of M.
fortunata, the edi"ce walls of EIFFEL TOWER were covered by Bathymodiolus azoricus. The Polynoidae
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1331
Fig. 3. Map of the LUCKY STRIKE vent "eld showing the location of the active sites (`EM 12a bathymetric survey
* FLORES cruise). The central topographic high is made up of a composite volcano 13 km long, 7 km wide and 430 m
high and divided into two parts separated by a N}S valley. The eastern part has a semicircular shape with three volcanic
cones at its summit. The LUCKY STRIKE vent "eld is situated around a circular lava lake between these three cones.
B. seepensis was present in almost all mussels collected. Mussel distribution at EIFFEL TOWER was of
particular note. A simple direct observation showed size segregation in the mussel beds (Comtet
and Desbruyères, 1998; Comtet, 1998) within the same site. In the samples collected on isolated
substrates (sulphide blocks and fragments) inside an active site, size spectrum analysis showed that
small individuals colonised less active zones, whereas large individuals were found in active areas,
on the smokers' walls, near the vent apertures. This is the reason why certain samples were
dominated (75%) by individuals belonging to the "rst cohort (mode 4.35 mm) whilst in others 63%
of the individuals belonged to older cohorts (modes 13}91 mm). Dense mats of "lamentous bacteria
covered certain areas of mussel beds of both sites. In the samples taken from mussel beds, several
accompanying species were found, in particular gastropods belonging to genera Protolira, Peltospira, Lepetodrilus and Shinkailepas and the amphipod Luckia striki. On the walls of small active
di!users of low and medium temperature (30}903C, e.g. in a small di!user), several tens of very
active individuals of C. chacei were observed. Numerous S. mesatlantica and pycnogonids were
present on the mussel beds, a large proportion of which was covered by "lamentous bacteria.
Numerous shrimps, mostly juveniles of M. fortunata or C. chacei, were aggregated around #anges
trapping hydrothermal #uid at the base of the edi"ce. C. chacei was also abundant among mussels.
1332
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Table 2
Equivalence between vent site names given during French and American cruises
Name
Statue of Liberty
Sintra
L'aiguille/M. Soares
Petit chimiste
Fanto( me
Ei!el Tower
Chimiste
Isabel
Markers PP
PP21
PP4
PP1
PP3}PP23
PP2
PP22
Markers Diva
Markers Florès
F15
F24
DII, IV, V, VI
DIII
DVIII
Markers US
US1
US3
F16
F22
F30
US6
US7
US4
MontseH gur
HeH lène
Pico
Nuno
Bairro Alto (ex. Pagodes)
Elisabeth
Y3
PP6
PP5
F17}F18
DX
PP7}PP8 PP25
PP12}PP13
PP24
White Castel
Crystal vent
F19
F29}F21
Jason
DI
F14
2608 site
In the crevices within the site perimeter, several "lter-feeding species were observed in low densities,
such as pedunculate cirripeds, as were small sessile carnivores such as Candelabrum phrygium
(hydroid). There was an abundant bathyal ichthyofauna around the sites, which make frequent
intrusions. The chimerid Hydrolagus pallidus was quite frequent; two or three individuals of
Cataetyx laticeps were always present at EIFFEL TOWER base as well as several Gaidropsarus n. sp.
living at a smoker's base inside crevices of the edi"ce. Lepidion schmidti was also frequently
observed (Saldanha and Biscoito, 1997, 2000).
The Y3 site is distinguishable from other LUCKY STRIKE sites by the occurrence of Rimicaris
exoculata in the upper and most active part of the edi"ce, mostly aggregated in crevices, when
mussels are concentrated at the base of the structure.
The RAINBOW vent "eld (36313N) is situated in the north AMAR segment (Fig. 4). It is the deepest
active segment of the ATJ area (2270}2320 m). The most active smokers are located at the western
and eastern ends of the hydrothermal "eld. The west zone has the roughest relief, while the sites of
the eastern zone are located on a sedimentary plateau. The highest developed sites are located at
the "eld centre (PP 28/35 and PP 29/37) and are either active edi"ces or nearly `extincta smokers.
Hydrothermal sediments border the vent "eld.
The youngest and most active sites (e.g. East sites and West sites) in this "eld were nearly azoic,
with the exception of the periphery of the western sites, where some sessile species were found.
Hydroids formed fairly dense covers over the surrounding hard substrates. In the eastern region,
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1333
Fig. 4. Map of the RAINBOW vent "eld. Left, the distribution of active sites. Right, topography of the overall area (EM 12
bathymetric survey * FLORES cruise). The Rainbow vent "eld is situated on the Rainbow Ridge in the north of AMAR
segment.
the polychaete Spiochaetopterus sp. formed dense and patchy populations inside hydrothermal
sediments near the site. On the active crest to the west of the "eld, M. fortunata was observed in
relatively dense aggregations along with some individuals of R. exoculata and C. chacei. No mussels
1334
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Table 3
Faunal composition of the vent communities at the three vent "elds (MENEZ GWEN, LUCKY STRIKE and RAINBOW)
Phylum
Porifera
Cnidaria
Annelida
Mollusca
Arthropoda
Genus
Hexactinellidae
Asbestopluma
Asbestopluma
Cladorhiza
Euchelipluma
Anthozoa
Stegolaria
Hydrozoa
Candelabrum
Eudendrium
Eudendrium
Cladocarpus
Polychaeta
Amathys
Archinome
Branchipolynoe
Capitella
Eunice
Cf. Hesiolyra
Levensteiniella
Lugia
Opisthotrochopodus
Spiochaetopterus
Prionospio
(Minuspio)
Prionospio
(Minuspio)
Prionospio
Monoplacophora Rokopella
Gastropoda
Alvania
Amphissa
Calliostoma
Dendronotus
Emarginula
Laeviphitus
Lepetodrilus
Lirapex
Mitrella
Neusas
Orbitestella
Orbitestellidae
n. gn.
Paraletopsis
Peltospira
Phymorhynchus
Protolira
Protolira
Pseudorimula
Pseudosetia
Shinkailepas
Strobiligera
Xylodiscula
Bivalvia
Bathymodiolus
Pycnogonida
Sericosura
Halacarida
Halacarellus
Copidognathus
Species
MG
LS
R
*
*
*
*
deH term.
*
*
*
*
*
Vacelet
id
id
id
Vervoort
id
Marques
Marques
Vervoort
Desbruyères
Kudenov
id
*
*
*
*
*
id
id
Pettibone
Desbruyères
Pettibone
sp.
n. sp.1
*
*
*
Petersen
Desbruyères
sp.2
*
*
id
sp.3
n. sp.
n. sp.
acuticostata
obesula
comteti
sp.
n. sp.
n. sp.
n. sp.
nitidunila
marshalli
n. sp.
n. sp.
*
pennatula
infundibulum
a!. grimaldi
pristina
geniculata
phrygium
sp.
rameum
formosus
lutzi
sp.
a!. seepensis
sp.
norvegica
sp.
n. sp.
sp.
n. sp.
n. sp.
n. sp.
n. sp.
valvatoides
thorvalldssoni
midatlantica
azorica
n. sp.
brychia
n. sp.
azoricus
heteroscela
alvinus
alvinus
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
id
WareH n & Bouchet
id
id
id
Valdes & Bouchet
WareH n & Bouchet
id
id
id
id
id
id
id
id
id
id
id
WareH n
WareH n & Bouchet
id
id
id
id
Von Cosel, Comtet et al.
Child & Segonzac
Bartsch
Bartsch
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1335
Table 3 (continued)
Phylum
Arthropoda
Cirripeda
Copepoda
Ostracoda
Amphipoda
Decapoda
Echinodermata
Chordata
Echinida
Ophiurida
Chondrichthyes
Osteichthyes
Genus
Species
MG
Altiverruca
Poecilasma
Poecilasma
Verum
Aphotopontius
Aphotopontius
Stygiopontius
Bathyconchoecia
Bairdia
Bythocypris
Krithe
? Pontocypris
Luckia
Bouvierella
Gitanopsis
Acanthephyra
Acanthephyra
Kemphyra
Nematocarcinus
Chaceon
Paromola
Segonzacia
Alvinocaris
Mirocaris
Chorocaris
Rimicaris
Echinus
longicarinata
aurantia
crassa
n. sp.
atlanteus
temperatus
rimivagus
pauluda
sp.
sp.
sp.
sp.
striki
curtimana
alvina
eximia
purpurea
sp.
exilis
aznis
cuvieri
mesatlantica
a!. markensis
fortunata
chacei
exoculata
alexandri
Ophioctenella
Apristurus
Etmopterus
Centroscymnus
Deania
Hydrolagus
Hydrolagus
Cataetyx
Chaunax
Chiasmodon
Coelorhynchus
Epigonus
Gaidropsarus
acies
maderensis
princeps
coelolepis
profundorum
aznis
pallidus
laticeps
sp.
niger
labiatus
telescopus
n. sp.
Halosaurus
Guttigadus
Lepidion
Lycenchelys
Mora
Nezumia
Polyacanthonotus
Simenchelys
Synaphobranchus
Trachyscorpia
johnsonianus ?
latifrons ?
schmidti
*
n. sp.
moro
*
sclerorhynchus
rissoanus
parasitica
*
kaupi
*
cristulata echinata*
LS
R
deH term.
*
Young
id
id
id
Humes & Segonzac
Humes
Humes & Segonzac
Carbonel
id
id
id
id
id
id
id
Biscoito
id
id
id
id
id
Guinot
Saint-Laurent
Segonzac
id
id
Sibuet
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
Tyler
Biscoito & Saldanha
id
id
id
id
id
id
id
id
id
id
Biscoito, Saldanha
& Desbruyères
id
id
id
Biscoito
Biscoito & Saldanha
id
id
id
id
id
1336
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
were observed around these sites except a few small individuals of Bathymodiolus azoricus around
the PP26 smoker. Two complex edi"ces (PP28/35 and PP37) were observed in the central part of
the "eld. They were composed of several active (2}3) and inactive (5}10) chimneys. Temperatures
measured at the di!user summit ranged from 22 to 633C. Low di!usion occurred along the lowest
part of some clogging chimneys and all around the site through the sediment. These edi"ces were
largely colonised by M. fortunata on sulphide di!users with higher densities on chimneys covered
by iron oxides (about twice as dense as on di!users). Dense swarms of R. exoculata were located in
small depressions between chimneys. At the PP37 site, a relatively large population of mussels had
grown on the areas of low di!usion and on clogging chimneys. Small-sized mussels were located at
the base of the chimneys where low di!usion occurred, while large-sized individuals were located
on the higher parts of the inactive chimneys covered by deposits. About 20% of the mussels
harboured the commensal polynoid B. seepensis. Along the active walls, a few bythograeid crabs
(S. mesatlantica) were observed, as well as some zoarcid "shes among mussels at the base of the site.
These proved to be a yet undescribed species of the genus Lycenchelys. Ophiurids, gastropods
(Phymorhynchus n. sp.) and pygnogonids also occured in this sediment covered area. Small
gastropods (Mitrella nitidunila, Protolira valvatoides, Lepetrodrilus n. sp., etc.), free polynoid and
ampharetid polychaetes were sampled at this site (Table 3).
3.2. Site characterisation
In order to characterise the di!erent microhabitats, we took a total of 70 discrete water samples
and recorded temperatures in the main macrofaunal assemblages and within di!use venting
structures in the three vent "elds. Within the LUCKY STRIKE vent "eld, three di!erent sites (EIFFEL
TOWER, BAIRRO ALTO and ELISABETH) were sampled. The averages and standard deviations of
temperature, pH and concentrations of chemicals that fuel bacterial autotrophy are shown in
Table 4. Cross comparisons between the #uids amongst habitats/sites were made using the Fisher
test for variances and t-Student's test for means of small samples when n54 (p"0.05). At the
RAINBOW site, the microhabitats of Rimicaris and Mirocaris did not di!er signi"cantly. The
microenvironment was signi"cantly di!erent for Rimicaris and Mirocaris vs. Bathymodiolus when
methane content, pH and temperature were compared. The LUCKY STRIKE site was characterised by
strong di!erences in the variances. The site-to-site comparison, concerning the mussels' microenvironment, revealed signi"cant di!erences between RAINBOW, EIFFEL TOWER (LUCKY STRIKE), BAIRRO
ALTO (LUCKY STRIKE) and MENEZ GWEN when pH was compared. Signi"cant di!erences were also
observed in the CO and methane concentrations between EIFFEL TOWER and BAIRRO ALTO on one
side and PP24 on the other. A decreasing gradient in S, CO , CH and temperature was found
when these parameters were measured at the di!users, in the Rimicaris and the Mirocaris habitats,
in the mixed habitat of Mirocaris and Bathymodiolus, in the mussel beds and the seawater. However,
when adjacent pairs of di!erent habitats were compared, this gradient could not discern the
microdistribution.
During the `VICTOR PREMIEE REa cruise, the use of the in situ auto-analyser `ALCHIMISTa
in the LUCKY STRIKE vent "eld allowed a better description of the vent population environment
(Fig. 5 and Table 5). Pro"les of temperature, sulphide and nitrate concentrations (Fig. 5) were
obtained as the probe and inlet were gradually moved over the di!erent animal populations within
a quadrat (0.25 m), terminating in an area of visible di!usion. The data obtained indicate general
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1337
Table 4
Mean values and standard deviations for chemical factors, temperature and pH as measured on discrete water samples at
RAINBOW, LUCKY STRIKE and MENEZ GWEN vent "elds. The data in bold type were not used in cross comparisons due
to the small number of sample replicates
Site
Sample
n
RAINBOW
PP 29
Di4user
2
Rimicaris
9
Mirocaris
5
Bathymodiolus
5
Di!user
5
Bathymodiolus
#Mirocaris
11
Bathymodiolus
5
Seawater
1
Di4user
2
Bathymodiolus
#Mirocaris
6
Bathymodiolus
6
Di4user
2
Bathymodiolus
ⴙMirocaris
1
Bathymodiolus
6
Bathymodiolus
4
LUCKY STRIKE
Ei!el Tower
LUCKY STRIKE
Bairro Alto
LUCKY STRIKE
Elisabeth
PP 24
MENEZ GWEN
T (3C)
pH
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
47.5
3.5
13.2
5.5
11.2
4.0
6.0
3.9
29.3
29.9
6.8
5.8
0.4
6.8
0.4
7.1
0.3
7.7
0.3
6.4
1.0
7.3
SD
Mean
SD
Mean
SD
Mean
SD
Mean
1.3
6.7
3.0
4.4
S (M)
CO (M)
CH (M)
NO\ (M)
12.8
13.4
8.5
6.1
7.2
8.7
3.8
4.1
18.8
30.6
2.4
3791
251
2046
208
2219
383
2169
233
408
2802
2332
112.4
31.4
21.9
25.6
6.5
4.3
0.2
0.1
61.9
115.2
0.8
15.4
0.4
15.5
3.6
17.1
0.7
16.7
1.9
16.3
3.0
19.7
0.4
7.1
0.2
7.79
2.4
2.1
1.4
238
2430
152
1.0
0.0
0.0
0.0
3.9
18.3
0.5
20.03
9.2
4.0
9.5
6.6
0.6
6.9
2.3
3.1
19.9
2209
401
2737
0.3
0.3
0.2
17.1
0.8
17.9
SD
Mean
SD
Mean
SD
Mean
SD
3.4
6.7
1.8
85.0
63.6
13.7
0.4
7.2
0.4
5.3
1.1
6.31
14.7
1.2
1.5
155.7
147.5
2.78
238
2472
276
2235
273
2748
0.2
0.1
0.1
4.8
6.4
0.0
0.5
18.1
1.0
11.8
7.3
Mean
SD
Mean
SD
5.7
0.9
10.1
0.5
7.6
0.2
6.7
0.2
1.3
0.7
0.7
0.4
2086
309
0.0
0.0
trends, as anticipated from discrete results (Table 5), but with a more precise de"nition of individual
habitats. The mussels belonging to the youngest cohorts (`small musselsa, shell length (30 mm)
were present in environments where the temperature was ca. 5.4}7.63C (Table 8) with small
temporal variations. In this very diluted environment sulphide concentrations ranged from 0.7 to
1 M. Cross comparisons (Fisher and t-Student's tests, p"0.05) were performed using the in situ
data obtained at the PP 24 vent site. The microhabitats described were signi"cantly di!erent from
each other for the three parameters measured, with a continuous increase of temperature and S
and a decrease in nitrate from the small mussels' habitat towards the di!users.
1338
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Fig. 5. Temperature (;), sulphide (䉭)and nitrate (*) concentrations experienced by di!erent populations of organisms
and recorded in situ by `ALCHIMISTa, in the LUCKY STRIKE vent "eld at Elisabeth PP24 site. `ALCHIMISTa is a chemical
in situ analyser, based on #ow-injection analysis and colorimetric detection.
Table 5
`ALCHIMISTa in situ temperature, sulphide and nitrate concentrations experienced by di!erent species in the LUCKY
STRIKE vent "eld at Elisabeth PP24 vent site
n
Small Bathymodiolus
Mean
SD
Large Bathymodiolus
Mean
SD
Mirocaris fortunata
Mean
SD
Di!user
Mean
SD
T (3C)
Nitrate (M)
Sulphide (M)
10
6.5
1.1
19.0
0.2
1.0
0.3
10.0
1.2
18.3
0.1
2.1
0.4
13.9
0.5
17.4
0.4
7
15.6
0.7
6.9
0.8
8
31.2
7.8
4
6
8
22.6
1.3
The metal concentrations amongst the di!erent microhabitats displayed high variations due to
the lack of distinction between particulate and dissolved metals. The mean values increase from
seawater to mussel#shrimps environment (Table 6). The metal concentrations are at least one
order of magnitude higher than those in seawater for Pb and Cu, two orders of magnitude higher
for Cu.
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1339
Table 6
Metallic concentrations (total dissolvable metals) experienced by di!erent communities as measured in discrete samples
Cu
g l\
Pb
g l\
Cd
g l\
N
Mean
SD
Mean
SD
Mean
SD
Sea water
Mussel beds
Mussel beds#shrimps
3
7.1
0.8
0.9
0.4
0.3
0.1
20
104.3
112.5
10.1
10.8
1.0
1.3
10
237.6
350.9
17.5
27.7
3.4
5.8
Table 7
Processed temperature time-series (HOBO威 and MICREL威 temperature recorders). Ambient temperatures were 3.73C
in RAINBOW, 4.53C in LUCKY STRIKE (L.S.) and 8.83C in MENEZ GWEN
Micro-habitats
Duration
Mean T
(3C)
RAINBOW
L.S./PP24
L.S./PP24
L.S./PP24
L.S./Ei!el Tower
L.S./Bairro Alto
L.S./Bairro Alto
L.S./Bairro Alto
Bathymodiolus
Small Bathymodiolus
Small Bathymodiolus
Bathymodiolus
Bathymodiolus
Bathymodiolus
Small Bathymodiolus
Bathymodiolus
and Mirocaris
Bathymodiolus
and Mirocaris
Bathymodiolus
and Mirocaris
Di!user
Di!user
Bathymodiolus
Bathymodiolus
11 months
11 months
11 months
11 months
1 month
5 days
5 days
5 days
3.7
5.5
5.6
12.8
6.4
5.2
6.6
8.2
0.5
0.7
1.0
3.7
0.8
0.7
0.9
1.7
6.3
9.6
16.7
25.5
10.5
8.4
9.1
14.2
3.1
4.2
4.6
4.8
4.5
3.9
4.0
4.6
3.2
5.4
12.1
20.7
6.0
4.5
5.1
9.6
5 days
11.4
2.6
19.1
4.1
15
5 days
14.4
3.0
24.6
6.0
18.6
5 days
5 days
1 day
1 day
21.3
38.8
9.6
9.8
2.8
3.8
0.4
0.5
29.9
49.5
11.7
12.1
6.3
26.8
8.3
8.8
23.6
22.7
3.4
3.3
L.S./Ei!el Tower
L.S./Ei!el Tower
L.S./Ei!el Tower
L.S./Bairro Alto
MENEZ GWEN
MENEZ GWEN
SD. T
(3C)
Max T
(3C)
Min T
(3C)
¹
Site
Temperature records covered a period ranging from a day to 11 months (Table 7). Temperatures
varied considerably during the recording period. All the habitats, even those characterised by
a mean temperature close to the ambient and by a reduced variability (e.g. mussel beds in RAINBOW
or small mussel patches in LUCKY STRIKE), a!orded signi"cant temperature bursts (e.g. respectively
6.3 and 16.43C). The comparison of the mean temperature values obtained during the "rst month
with those obtained during the last month (11 months interval) showed a signi"cant change. At
PP24, a 5.93C decrease in temperature was recorded among mussels. Inside large-sized mussel
1340
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Table 8
Elemental composition of particles collected in sediment traps at RAINBOW, LUCKY STRIKE, MENEZ GWEN and o!-vent
site (pelagic #ux)
Mean total #ux (g m\ d\)
Organic carbon (%)
Inorganic carbon (%)
Organic carbon #ux (mg m\ d\)
Nitrogen (%)
Sulphur (%)
Calcium (%)
Iron (%)
Copper (%)
Barium (%)
Rainbow
Lucky Strike
Menez-Gwen
Pelagic #ux
6.9
0.136
Not detected
9.38
0.014
17.24
26.6
5.97
3.59
0.25
0.26
3.94
3.48
10.41
0.71
10.76
10
6.3
0.93
3.19
0.64
1.62
0.98
10.36
0.12
18.17
26.2
1.3
0.5 detected
1.7 detected
0.032
10.8
8.0
3.47
1.45
0.54
*
*
*
*
clumps at LUCKY STRIKE vent "eld, temperatures ranged from ambient to 25.53C, i.e. the mean
temperature of some di!users.
Mean particle #ux was 6.9 g m\ d\ at RAINBOW and 0.26 g m\ d\ at LUCKY STRIKE. At the
MENEZ GWEN station, the trap on the mooring close to the vent sampled a steady #ux of
0.64 g m\ d\ over 10 days, and no di!erence was observed between the two sampling periods
(5 days each). At the same time, the pelagic #ux measured with the trap positioned o! MENEZ GWEN
hydrothermal "eld was 0.032 g m\ d\. The main feature of the hydrothermal vent particles
(Table 8) was the high concentration of sulphur (10.7}18.2%) in comparison with that of pelagic
particles (0.5%). The organic carbon concentration was particularly low (0.14%) at RAINBOW.
Barium concentration was high at MENEZ GWEN and Lucky Strike, whereas copper predominated
at RAINBOW.
4. Discussion
Biogeography is in#uenced by local ecological and historical factors. Concerning the latter,
hydrothermal-linked population homogeneity along an active ridge axis is conditioned by the
presence or absence of `relaya sites between two "elds and of major discontinuities (transform
faults or continental plates). Distances between active zones are greater at slow spreading ridges
than at fast ones, and thus exchanges between populations may be less frequent at the former.
A linear model correlating the increase in the number of hydrothermal plumes with the increase of
a ridge spreading rate has been proposed by Baker et al. (1995) and was recently re-examined
following the discovery of many hydrothermal plumes along the South Western Indian Ridge, an
ultra slow spreading ridge (Van Dover, 1998). German et al. (1996) have mapped seven hydrothermal plumes over 200 km of the Mid-Atlantic Ridge between 36 and 383N. However, an unequivocal
relationship between the presence of a plume and the existence of an active "eld could not be
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1341
established. In the present case (slow spreading ridge) distances between known "elds in this area
are relatively small (89 km between MENEZ GWEN and LUCKY STRIKE, 183 km between LUCKY STRIKE
and RAINBOW) and therefore not in agreement with the above-mentioned model.
The "rst observations made at the shallower "elds (850}1700 m) of the Azores Triple Junction,
LUCKY STRIKE (Van Dover et al., 1993) and MENEZ GWEN (Desbruyères et al., 1994), contrast with
those made on the southern and deeper "elds, SNAKE PIT (233N}3480 m) and TAG (263N, 3650 m),
as the former are dominated by mussels (Bathymodiolus azoricus) and the latter by shrimps, in
particular R. exoculata (Rona et al., 1986; Galkin and Moskalev, 1990; Gebruk et al., 1993;
Segonzac, 1992). This di!erence suggested the existence of isolation mechanisms, which could be
either hydrostatic pressure thresholds or transform faults and fracture zones, and therefore the
presence of two biogeographical provinces along the Mid-Atlantic Ridge, as proposed by Van
Dover et al. (1993) and Desbruyères et al. (1994). Furthermore, the description of the "elds of
Broken Spur (293N, 3000 m) (Murton et al., 1995; Copley et al., 1997) and Logatchev (14345N;
3000 m) (Gebruk et al., 1997) showed associations dominated by R. exoculata (Shank et al., 1998),
thus con"rming the dominance of shrimps in the "elds south of the Atlantis Fracture Zone, deeper
than 3000 m. Finally, the discovery of RAINBOW hydrothermal vent "eld at an intermediate latitude
and depth (363N, 2300 m) showed a situation more similar to that found at SNAKE PIT and BROKEN
SPUR, with shrimps, especially R. exoculata and M. fortunata, present in great abundance at all active
sites of the "eld. This intermediate situation, along depth and latitude gradients, is not in agreement
with the two-provinces hypothesis referred to above.
The mussels present at RAINBOW vents belong to the same species as those at LUCKY STRIKE and
MENEZ GWEN. Even though mussel populations are, at certain sites, rather abundant, they are
located at the base of the edi"ces, as in the case of SNAKE PIT and rarely accommodate small
individuals. Considering the dominant species and their relative contribution to the population
composition, LUCKY STRIKE and MENEZ GWEN can be separated from other Mid-Atlantic Ridge
"elds. Moreover, taking all presently identi"ed species and in spite of an apparent similarity
between LUCKY STRIKE and MENEZ GWEN populations, the Jaccard index gives a low speci"c
similarity (ca. 20% between MENEZ GWEN and LUCKY STRIKE). A similarity of the same magnitude is
obtained between RAINBOW and MENEZ GWEN (20%) and is slightly higher between RAINBOW and
LUCKY STRIKE (27%). Also, the speci"c similarity between RAINBOW and SNAKE PIT is low (22%) in
spite of the general resemblance of both "elds' communities. There is a remarkable di!erence in
terms of similarity between "elds, when the Atlantic is compared with the EPR. A similarity of over
40% is found between 13 and 213N on the EPR. These two "elds are ca. 1000 km away from each
other and are separated by the Rivera and Orozco transform faults, but their depths (ca. 2600 m)
are similar.
The main faunal di!erences observed between LUCKY STRIKE and MENEZ GWEN hydrothermal
"elds are due to a depletion of hydrothermal endemic species (respectively, 28 and eight endemic
species) and to the penetration of bathyal species (40% in Rainbow, 72% in Menez Gwen), and
therefore not to vicariance events. The populations of these two sites and those of RAINBOW
are without doubt di!erent because of a response to environmental conditions rather than
geographic isolation mechanisms (Pico fault between LUCKY STRIKE and RAINBOW causes a 45-km
long thrust).
In addition to the direct biological impact of hydrostatic pressure, bathymetric variations
linked to the Azores hotspot presence produce a progressive change of the micro-hydrothermal
1342
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
environment, not only among di!erent "elds, but also among sites within the same "eld, as in the
case of LUCKY STRIKE vent "eld, under the action of phase separation phenomena. Additional
chemical variations may also be associated with the presence of ultrama"c rocks at RAINBOW.
Nevertheless, in the same habitats in di!erent vent "elds, nutrient (H S and CH ) and metallic
ion concentrations and temperatures weakly vary, probably because of the organisms' speci"c
requirements.
Within the vent "elds, chemical analyses of discrete #uid samples were unable to signi"cantly
discriminate between the environments surrounding di!erent shrimp species and mussels (except
for the mussels at RAINBOW vent "eld). Two biases could jointly explain this inability: (1) mixing
of microenvironments during sampling, (2) inadequacy of localisation of sample vs organism
micro-habitat. Nevertheless, this set of data (plus the temperature time series) emphasises the
variability of the microenvironment and the possible overlap between spatial distributions of
individual species, and suggests the importance of inter-speci"c competition in the structuring of
vent communities. In situ chemical analyses made at LUCKY STRIKE vent "eld (Ei!el Tower site)
showed that a physical and chemical gradient does exist within the vent habitat. Caridean
shrimps are the most eulithophilic organisms (Fustec et al., 1987), and adult mussel beds colonise
the most diluted part of the gradient with the young individual clusters at the periphery (Sarradin
et al., 1999). In RAINBOW vent "eld and Y3 (LUCKY STRIKE) segregation between the habitat of
R. exoculata and that of mussels was also observed. M. fortunata and C. chacei were found to have
a rather opportunistic distribution across the whole gradient. We can hypothesise that the presence
of dense mussel beds covering the chimney walls at LUCKY STRIKE vent sites, which change the
microenvironment (Johnson et al., 1994) and screen the sulphide wall, may preclude the development of R. exoculata swarms. This hypothesis must be tested in the future by experimental
exclusion.
Even if Bathymodiolus azoricus harbours chemoautotrophic endosymbionts, it still has a "ltering
behaviour, not only for feeding from particulate organic matter (Le Pennec and Hily, 1984; Le
Pennec and Prieur, 1984; Le Pennec et al., 1990), but also for providing its symbionts with methane
or hydrogen sulphide and for respiration. The very high production of particles at Rainbow could
negatively a!ect their "ltering activity and then explain their relative exclusion from this "eld.
When fed at high concentration of particles of low organic content, the shallow-water mytilids
Mytilus edulis and M. trossulus can increase their "ltration rate, reject a higher proportion of "ltered
material as pseudofaeces, and select organic-rich particles, thus maximising organic matter ingestion (Ari"n and Bendell-Young, 1997; Bayne, 1976). However, the e$ciency of such processes varies
among bivalve species, and the feeding activity can be altered by high concentrations of inorganic
particles, as in Placopecten magellanicus and Mercenaria mercenaria (Bricelj and Malouf, 1984;
Cranford and Gordon, 1992). High particle concentrations could also a!ect juveniles, as Bricelj
et al. (1984) demonstrated for Mercenaria mercenaria. Then we cannot reject the hypothesis that
high concentrations of vent particles have a negative e!ect on Bathymodiolus, although the
mechanisms are unknown. When particle concentration in the #uid decreases, "ltering symbiotic
species become dominant (LUCKY STRIKE and MENEZ GWEN vent "elds).
Populations in the two shallower ATJ "elds di!er from those at other Mid-Atlantic Ridge sites,
mainly because of the dominance of Bathymodiolus azoricus. RAINBOW "eld communities' composition and biodiversity, in spite of depth (2320 m) and the existence of transform faults (Haynes,
Atlantis and Kane and Fifteen Twenty), are closer in structure to those of TAG, BROKEN SPUR and
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1343
SNAKE PIT. Chemical di!erences of the #uids surrounding organisms associated with the ultrama"c
rocks at the hydrothermal sites seem not to produce major faunal modi"cations. When particle
concentration in the #uid decreases, "ltering symbiotic species become dominant in relation to
particle feeders and bathyal species penetrate in active areas to feed. The supposed time and
space stability of sites at slow spreading ridges and the relative fragmentation of the environment give place to a weak speci"c similarity between sites, in terms of accompanying species.
When comparing communities and environments at the three-studied vent "elds on the
ATJ, one cannot consider the existence of a single biogeographical province but rather a succession of several distinct biogeographic islands having di!erent associations and habitats. Of
course, one cannot rule out the existence of a direct e!ect of the pressure on the biological
associations and metabolisms, although a large proportion of the bathyal species is known to be
eurybathic between 800 and 2500 m. More likely, the hydrostatic variation leads to a rapidly
varying habitat under phase separation processes within that bathymetric zone (ecotone patterns).
At the shallowest vents, the hydrothermal #uid loses partly its toxic substances (dissolved and
particulate metals and sulphide) but also a part of the nutrients fuelling the chemoautotrophic
processes. We hypothesise that the decrease in #uid toxicity and physical aggressiveness
allow `non-vent predatorsa to enter the hydrothermalism-a!ected area, to feed on remaining
endemic species. This hypothesis may account for the decrease in number of non-endemic species
with depth.
Acknowledgements
We thank the crews of N/O Nadir, N/O l+Atalante, R/V Arquipe& lago and manned submersible
Nautile for invaluable technical assistance with deployment and retrieval of experimental arrays
and the collection of specimens. We are indebted to the taxonomists who informed us of the
content of pending manuscripts (R. Von Cosel, A. WareH n and P. Bouchet), and to J.G. Quinn for
revising the text. We appreciate the helpful comments of C.L. Van Dover and one anonymous
reviewer on the manuscript. This research has been undertaken in the framework of the AMORES
project. We acknowledge the support from the European Commission's Marine Science and
Technology Program (MAST III) under contract MAS3-CT95-0040. It was also partly funded by
Fundac7 a o para a Cie( ncia e Tecnologia, Programa Praxis XXI-Portugal. The cruises DIVA 2,
MARVEL & PICO were supported by IFREMER, France.
References
Ari"n, Z., Bendell-Young, L.I., 1997. Feeding response and carbon assimilation by the blue mussel Mytilus trossulus
exposed to environmentally relevant seston matrices. Marine Ecology Progress Series 160, 241}253.
Baker, E.T., German, C.R., Elder"eld, H., 1995. Hydrothermal plumes over spreading-center axes: global
distributions and geological inferences. In: Humphris, S.E., Zierenberg, R.A., Mullineaux, L.S., Thomson, R.E.
(Eds.), Sea#oor Hydrothermal Systems, Vol. 91. American Geophysical Union Publishing, WA,
pp. 47}71.
Bayne, B.L., 1976. Marine Mussels: their Ecology and Physiology. Cambridge University Press, London.
1344
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Bricelj, V.M., Malouf, R.E., 1984. In#uence of algal and suspended sediment concentrations on the feeding physiology of
the hard clam Mercenaria mercenaria. Marine Biology 84, 155}165.
Bricelj, V.M., Malouf, R.E., Quillfeldt, C., 1984. Growth of juvenile Mercenaria mercenaria and the e!ect of resuspended
bottom sediments. Marine Biology 84, 167}173.
Cann, J., Van Dover, C., Walker, C., Dando, P., Murton, B., 1994. Diversity of vent ecosystems (DOVE). Bridge
Newsletter 4, 1}39.
Charlou, J.-L., Donval, J.-P., Douville, E., Knoery, J., Fouquet, Y., 2000. Geochemistry of hydrothermal #uids collected
at Lucky Strike (37317N) and Menez Gwen (37350N) hydrothermal "elds, South of the Azores Triple Junction on the
Mid-Atlantic Ridge (Diva 1 cruise, May 1994). Chemical Geology, in press.
Charlou, J.-L., Donval, J.-P., Douville, E., Knoery, J., Fouquet, Y., Bougault, H., Jean-Baptiste, P., Stievenard, M.,
German, C., 1997. High methane #ux between 153N and the Azores Triple Junction, Mid-Atlantic Ridge. Hydrothermal and serpentinization processes. San Francisco, CA, USA, American Geophysical Union, Fall meeting,
831.
Charlou, J.L., Donval J.P., Fouquet, Y., Douville, E., Knoery, J., Jean-Baptiste, P., Dapoigny, A., Stievenard, M., 2000b.
Geochemistry of hydrothermal #uids collected at Lucky Strike (37317N) and Menez Gwen (37350N) hydrothermal
"elds, south of the Azores Triple Junction on the Mid-Atlantic Ridge (DIVA 1 cruise * May 1994). Journal of
Geophysical Research, submitted for publication.
ChevaldonneH , P., Jollivet, D., Vangriesheim, A., Desbruyères, D., 1997. Hydrothermal-vent alvinellid polychaete dispersal
in the eastern Paci"c. 1. In#uence of vent site distribution, bottom currents, and biological patterns. Limnology and
Oceanography 42, 67}80.
Comtet, T., 1998. Structure des populations, reproduction, croissance et phylogeH ographie des Mytilidae des champs
hydrothermaux Lucky Strike et Menez Gwen (37317N et 37350N sur la dorsale meH dio-atlantique). Thèse Doctorale
en OceH anographie Biologique. Brest, France, UniversiteH de Bretagne Occidentale, 244pp.
Comtet, T., Desbruyères, D., 1998. Population structure and recruitment in mytilid bivalves from the Lucky Strike and
Menez Gwen hydrothermal vent "elds (37317N and 37350N on the Mid-Atlantic Ridge). Marine Ecology Progress
Series 163, 165}177.
Copley, J.T.P., Tyler, P.A., Murton, B.J., Van Dover, C.L., 1997. Spatial and interannual variation in the faunal
distribution at Broken Spur vent "eld (293N, Mid-Atlantic Ridge). Marine Biology 129, 723}733.
Cranford, P.J., Gordon, D.C., 1992. The in#uence of dilute clay suspensions on sea scallop (Placopecten magellanicus)
feeding activity and tissue growth. Netherlands Journal of Sea Research 30, 107}120.
Desbruyères, D., Alayse-Danet, A.-M., Ohta, S., STARMER and BIOLAU cruises participants, 1994. Deep-sea
hydrothermal communities in Southern Paci"c back-arc basins (the North Fiji and Lau Basins): composition,
microdistribution and food-web. Marine Geology 116, 227}242.
Fonselius, S.H., 1983. Determination of hydrogen sul"de. In: Grasshof, K. (Ed.), Methods of Sea-water Analysis. Verlag
Chemie, pp. 73}84.
Fornari, D., Van Dover, C.L., Shank, T., Lutz, R., Olsson, M., 1994. A versatile, low-cost temperature sensing device for
time-series measurements at deep sea hydrothermal vents. Bridge Newsletter 6, 40}47.
Fouquet, Y., Charlou, J.-L., Donval, J.-P., Radford-Knoery, J., OndreH as, H., Cambon, P., Bougault, H., Etoubleau, J.,
Barriga, F.J.A., Costa, I.M.A., Lourenc7 o, N., Tivey, M.K., 2000. Hydrothermal processes on shallow volcanic
segments: Mid Atlantic Ridge near the Azores Triple Junction. Earth and Planetary Science Letters,
submitted.
Fustec, A., Desbruyères, D., Juniper, S.K., 1987. Deep-sea hydrothermal vent communities at 133N on the East Paci"c
Rise: microdistribution and temporal variations. Biological Oceanography 4, 121}164.
Galkin, S.V., Moskalev, L.I., 1990. Hydrothermal fauna of the Mid-Atlantic Ridge. Oceanology 30, 624}627.
Gebruk, A.V., Moskalev, L.I., ChevaldonneH , P., Sudarikov, S.M., Chernyaev, P., 1997. Hydrothermal vent fauna
of the Logatchev area (14345N, MAR): preliminary results from "rst Mir and Nautile dives in 1995. InterRidge News
6, 1}8.
Gebruk, A.V., Pimenov, N.V., Savvichev, A.S., 1993. Feeding specialization of bresiliid shrimps in the TAG site
hydrothermal community. Marine Ecology Progress Series 98, 247}253.
German, C., Parson, L.M., Heat Scienti"c Team, 1996. Hydrothermal exploration near the Azores Triple
Junction: Tectonic control of venting at slow-spreading ridges. Earth and Planetary Science Letters 138,
93}104.
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
1345
Hessler, R.R., Lonsdale, P.F., 1991. Biogeography of Mariana Trough hydrothermal vent communities. Deep-Sea
Research 38, 185}199.
Johnson, K.S., Childress, J.J., Beehler, C.L., Sakamoto, C.M., 1994. Biochemistry of hydrothermal vent mussel communities: the deep-sea analogue to the intertidal zone. Deep-Sea Research 41 (7), 993}1011.
Johnson, K.S., Childress, J.J., Hessler, R.R., Sakamoto-Arnold, G.M., Beehler, C.L., 1988. Chemical and biological
interactions in the Rose Garden eastern Paci"c Ocean hydrothermal vent "eld Galapagos spreading center. Deep-Sea
Research, Part A 35, 1723}1744.
Khripouno!, A., AlbeH ric, P., 1991. Settling of particles in a hydrothermal vent "eld (East Paci"c Rise 133N) measured
with sediment traps. Deep-Sea Research 38, 729}744.
Le Bris, N., Sarradin, P.-M., Alayse, A.-M., 2000. First application of a new submersible chemical analyser for deep
environment studies: in situ nitrate and total sul"de measurements at a 1600 m depth hydrothermal site. Marine
Chemistry, 72, 1}15.
Le Pennec, M., Donval, A., Herry, A., 1990. Nutritional strategies of the hydrothermal ecosystem bivalves. Progress in
Oceanography 24, 71}80.
Le Pennec, M., Hily, A., 1984. Anatomie, structure et ultrastructure de la branchie d'un Mytilidae des sites hydrothermaux du Paci"que oriental. Oceanologica Acta 7, 517}523.
Le Pennec, M., Prieur, D., 1984. Observations sur la nutrition d'un Mytilidae d'un site hydrothermal actif de
la dorsale du Paci"que oriental. Comptes Rendus de l'AcadeH mie des Sciences de Paris, SeH rie III 298,
493}498.
Lonsdale, P.F., 1977. Clustering of suspension-feeding macrobenthos near abyssal hydrothermal vents at oceanic
spreading centers. Deep-Sea Research 24, 857}863.
Murton, B.J., Van Dover, C., Southward, E., 1995. Geological setting and ecology of the Broken Spur hydrothermal vent
"eld: 29310N on the. In: Parson, L.M., Walker, C.L., Dixon, D.R. (Eds.), Hydrothermal Vents and Processes, Vol. 87.
The Geological Society of London Special Publication, pp. 33}41.
Riso, R.D., Monbet, P., Le Corre, P., 1997. Measurement of copper in sea-water by constant current stripping analysis
(CCSA) with a rotating gold disk electrode. The Analyst 122, 1593}1596.
Rona, P.A., Klinkhammer, G., Nelsen, T.A., Trefry, J.H., Elder"eld, H., 1986. Black smokers, massive sulphides and vent
biota at the Mid-Atlantic Ridge. Nature 321, 33}37.
Saldanha, L., Biscoito, M., 1997. Fishes from the Lucky Strike and Menez Gwen hydrothermal vent sites (Mid-Atlantic
Ridge). Boletim do Museu Municipal do Funchal 49, 189}206.
Saldanha, L., Biscoito, M., 2000. Gaidropsarus mauli a new species of Gadid "sh from the Lucky Strike hydrothermal vent
site (mid-Atlantic Ridge) and the Biscay Slope. Copeia, submitted for publication.
Sarradin, P.-M., Caprais, J.-C., 1996. Analysis of dissolved gases by headspace sampling gas
chromatography with column and detector switching. Preliminary results. Analytical Communications 33,
371}373.
Sarradin, P.-M., Caprais, J.-C., Riso, R., KeH rouel, R., Aminot, A., 1999. Chemical environment of the hydrothermal
mussel communities in the Lucky Strike and Menez Gwen vent "elds, Mid Atlantic Ridge. Cahiers de Biologie Marine
40, 93}104.
Segonzac, M., 1992. Les peuplements associeH s à l'hydrothermalisme oceH anique du Snake Pit (dorsale meH dio-atlantique;
233N, 3480 m): composition et microdistribution de la meH gafaune. Comptes Rendus de l'AcadeH mie des Sciences de
Paris, SeH rie III 314, 593}600.
Shank, T.M., Lutz, R.A., Vrijenhoek, R.C., 1998,. Molecular systematics of shrimp (Decapoda: Bresiliidae) from deep-sea
hydrothermal vents, I: Enigmatic `small orangea shrimp from the Mid-Atlantic Ridge are juvenile. Molecular Marine
Biology and Biotechnology 7 (2), 88}96.
Tunnicli!e, V., 1991. The biology of hydrothermal vents: ecology and evolution. Oceanography and Marine Biology
Annual Review 29, 319}407.
Van Dover, C.L., 1995. Ecology of Mid-Atlantic Ridge hydrothermal vents. In: Parson, L.M., Walker, C.L., Dixon, D.R.
(Eds.), Hydrothermal Vents and Processes, Vol. 87. The Geological Society of London Special Publication, pp.
257}294.
Van Dover, C.L., 1998. Vents at higher frequency. Nature 395, 437}438.
Van Dover, C.L., Desbruyères, D., Saldanha, L., Fiala-Medioni, A., Langmuir, C., 1993. A new faunal province at the
Lucky Strike hydrothermal vent "eld. Eos 74, 100.
1346
D. Desbruye% res et al. / Deep-Sea Research I 48 (2001) 1325}1346
Von Damm, K.L., Grant, B., Edmond, J.M., 1983. Preliminary report on the chemistry of hydrothermal solutions at 213
North, East Paci"c Rise. In: Rona, P.A., BostroK m, K., Laubier, L., Smith Jr., K.L. (Eds.), Hydrothermal Processes at
Sea#oor Spreading Centers. Plenum Press, New York, pp. 369}390.
Weliky, K., Suess, E., Ungere, C., Muller, P., Fisher, K., 1983. Problems with accurate carbon measurements in
marine sediments and water column particulates: a new approach. Limnology and Oceanography 28,
1252}1259.